Part 1: Evaluation of Mini-Split Heat Pumps as Supplemental and Full System Retrofits in a Hot-Humid Climate

November 30, 2017
Winter 2017
A version of this article appears in the Winter 2017 issue of Home Energy Magazine.
Click here to read more articles about Heating

More stringent energy efficiency standards and the demand for higher efficiency are increasing the popularity of technologies such as mini-split heat pumps (MSHPs). The market for ductless heat pumps is growing 10–30% annually. The U.S. mini-split market has seen a 221% growth rate over the last seven years, and a 20% growth rate is expected to continue, thanks partly to the introduction of the highly efficient, inverter-driven compressor. Growth projections are supported by high customer satisfaction, but most of the research has been conducted on space-heating applications.

However, studies report highly variable MSHP energy savings results. In heating-dominated climates, MSHPs are typically supplemental retrofit installations, providing the primary source of heating to the rooms that are most used. Research on southern-climate MSHP installations, where the primary focus is on cooling, is limited. However, the sparse data suggest that MSHPs are installed primarily to replace noisy window units or to serve a previously unconditioned space, rather than to displace less-efficient central air-conditioning systems, where they can also be useful. MSHP research needs to focus more on field performance, as opposed to laboratory testing; on measured performance and savings evaluations in different climates zones; and on submetered load shape, energy use, and energy savings information. A case study evaluating MSHPs as a complete system solution in the hot-humid climate found comfort issues, and recommended that future research focus on field performance data.

The project described in this article was conducted to address the need for this research. The Phased Deep Retrofit (PDR) project, conducted by the DOE Building America team Partnership for Improved Residential Construction (BAPIRC) in collaboration with Florida Power & Light (FPL), sought to determine what impact the installation of advanced residential technologies had on annual energy reductions. Using this platform, researchers investigated the impact of MSHP retrofits to answer questions about energy use savings, peak load shaving, and comfort issues in the hot-humid climate.


Total house power as well as detailed energy end-use data were collected to evaluate energy reductions and the economics of each retrofit at each PDR study home. All of the homes were audited and instrumented during the second half of 2012; shallow retrofits were conducted from March 1 through June 30, 2013. Monitoring of hourly house power and the various end uses was accomplished with a 24-channel data logger, supplemented by portable loggers to take temperature and relative humidity (RH), located near the central system thermostat. (Additional portable loggers were launched to obtain room-by-room comfort data). Hourly power was measured by SiteSage loggers, generally using 50 amp current transformers. Interior temperatures were measured near the thermostat using Onset HOBO U-10-003 portable loggers.

Evaluation Method

Regression analysis was used to project space-conditioning energy savings for each retrofit measure. To estimate pre- and postretrofit annual space-heating and -cooling energy use, daily average ambient temperatures were regressed against monitored daily HVAC energy use.

In keeping with the statistical analytical concept of parsimony, this study used the simplest model that showed stable and reliable results with strong explanatory power, a linear regression using a single independent variable. Outdoor temperature (in degrees Fahrenheit) was used rather than outdoor temperature minus indoor temperature, because of expected behavioral changes with the supplemental or total MSHP replacement. Differences in interior temperature are likely with the MSHP, because uniform interior room temperatures do not typically yield the greatest comfort. Brand found that space-conditioning systems that facilitate zoning have significantly lower energy use. When a supplemental or total MSHP is added, it becomes likely that occupants will maintain different heating and cooling conditions in different rooms of the home.

For each site, the typical meteorological year 3 (TMY3) weather data were applied to the regression coefficients to normalize the savings.

Supplemental Ductless Mini-Split Heat Pump Measure

Ten central Florida project homes received supplemental MSHPs from August 27, 2014 to July 23, 2015. The research question was Can a very efficient ductless mini-split heat pump be added centrally to homes already possessing a conventional central ducted system to reduce the run time of the lower-efficiency central system?

The equipment chosen for this measure was the 1-ton ductless, inverter-driven Panasonic XE12PKUA. This is a 25.5 SEER unit with a heating seasonal performance factor (HSPF) of 12. The variable-speed units have an Air-Conditioning, Heating, and Refrigeration Institute (AHRI)-rated nominal cooling capacity of 11,580 Btu per hour at the 95/80/67 rating condition, ranging from 2,800 to 14,000 Btu per hour; and a heating capacity of 13,800 Btu per hour at 47°F outdoors. As shown in Table 1, homes receiving the supplemental MSHPs were highly varied, with central air-conditioning systems of various ages and efficiencies. Duct systems for all central systems were located in the attic space of each home.

Table 1. Supplemental MSHP Site and Existing HVAC Characteristics

a Air Handling Unit
b Some systems were apparently unmatched; stated are manufacturer listed compressor efficiencies.
c Duct leakage measured at a test pressure of negative 25 pascals with respect to the outside, divided by the building’s conditioned floor area. Site 12 results were projected as test pressures could not be reached.

The supplemental MSHPs were expected to reduce space-cooling and -heating energy by reducing the run time of less-efficient existing central systems subject to duct losses. However, the configuration with two different systems with potentially competing thermostats serving a single zone made it hard to guess how this would work out. Moreover, no existing simulation model can provide savings estimates, because having two HVAC systems serve the same zone violates limits for hourly calculations.

The indoor fan coil was centrally located within each home. In most cases, the unit was located as close as possible to the central return grille of the existing system, to help with room-to-room distribution of MSHP air when the central unit was operating. At install, the cooling set point of the MSHP was set either 2°F or 4°F lower than the central system temperature. Owners were advised against setting the central system fan to run constantly, to prevent moist air from compressor coils from being brought back into the home, and to avoid excessive fan energy consumption.

Supplemental MSHP: Results and Discussion

To examine how the supplemental MSHP influenced space-cooling and -heating energy, an evaluation method was applied to the measured data for each installation. The evaluation periods generally spanned about 18 months in total: January 2014 through July 2015 for the late-summer 2014 installations, and summer 2014 through January 2016 for the early-summer 2015 installations. Evaluation periods varied according to installation date, other retrofit measures, and other significant changes, such as a change in occupancy. Energy savings results were normalized to TMY3. Tables 2 and 3 show the cooling and heating results from the regressions ,along with the hourly-logged interior temperature (Int T) and RH (for cooling), as measured near the thermostat, before and after the MSHP retrofit. Space- cooling energy savings were large, with a median of 33%, 2,007 kWh per year, or about 6.7 kWh per day. The median daily space-heating savings were 59%, 390 kWh per year or about 6.5 kWh per day.

Table 2. TMY3-Normalized Cooling Energy Use and Savings from the Supplemental MSHPs

Table 2. TMY3-Normalized Cooling Energy Use and Savings from the Supplemental MSHPs

Table 3. TMY3-Normalized Heating Energy Use and Savings from the Supplemental MSHPs

Table 3. TMY3-Normalized Heating Energy Use and Savings from the Supplemental MSHPs
b Site had no pre-retrofit heating to measure.

Examining the sites as a group, cooling-season interior temperature and RH were relatively similar between pre and post periods, on average, and with dew points averaging 68.3°F pre and 69.2°F post. On average, no moisture removal advantage of the supplemental MSHP was apparent. The heating-season interior temperature was also essentially unchanged between periods, on average; however, there were some increases. Cooling-season ambient temperature was typically slightly cooler, averaging 0.4°F lower postretrofit (76.9°F versus 76.5°F).

However, there were significant variations for some homes. For instance, the innovative configuration showed a large improvement to interior temperature and RH at Site 24, with reductions of 1.7°F and 6.5%, respectively, where the average dew point exceeded 69°F both pre- and postretrofit.

The impact on savings can be large.

Assuming an 11.5% increase in energy use for every 1°F of postretrofit take-back, cooling energy savings would have been about 18% without the take-back.

The average percentage space-heating energy savings achieved by the supplemental MSHPs was greater than the average percentage cooling energy savings. This was because Sites 16, 21, 23, 24, 27, and 60 had electric- resistance heating (the other four sites had heat pumps). The MSHPs with the much more efficient, inverter-driven heat pump technology provided most of the heating capacity, which eliminated or reduced auxiliary strip heat of central systems. Figure 1 shows the time series data, where electric-resistance strip heat is highly visible, as is the reduction to the space cooling in summer and the very low power of the MSHP system for Site 60.

Supplemental MSHP Performance on Peak Summer Days

Site 60 HVAC Energy Use Pre- and Post-Supplemental MSHP Installation
Figure 1. Time series data showing heating, ventilating, and air-conditioning energy use by air-conditioner compressor, air handler unit and strip heat, and supplemental mini-split for Site 60.

The projected HVAC annual energy savings from the supplemental MSHP measure for all ten sites averaged 34% or 2,357 kWh per year. Results have been normalized using population weighted TMY3 weather stations to represent average savings estimates for the FPL service territory. Table 4 summarizes the projected annual savings.

Table 4. TMY3-Normalzed Annual Cooling and Heating Energy Use and Savings from the Supplemental Mini-Splits

Table 4. TMY3-Normalzed Annual Cooling and Heating Energy Use and Savings from the Supplemental Mini-Splits

The average full retail cost for equipment, materials, and labor for each of the ten supplemental MSHP installations was about $3,900. The median annual HVAC energy savings translates into about $285 saved per year (2,375 kWh/year * 0.12/kWh), which yields a simple payback in about 14 years and an annual rate of return of 7.3%. In a mature market, economics are likely to improve with equipment and labor cost reductions—particularly as crews become more familiar with the relatively simple job of installing MSHPs. This cost analysis does not consider one notable benefit to the consumer—the redundant heating-and-cooling system for the home, which is highly desirable given the inevitable and unpredictable failure of central air-conditioning systems, some of which may take a few days to repair.

In order to evaluate the effect of the supplemental MSHPs during peak summer and winter hours, HVAC power demand at the utility coincident peak hours in 2014 were compared to that of 2015. Figure 2 compares the average HVAC demand of the ten supplemental MSHP sites for the summer peak, showing a large demand reduction of 0.50 kWh, or 16%. The winter peak evaluation is limited to the six supplemental MSHPs installed in 2014. Figure 3 compares the average HVAC demand at these sites, which shows a very large demand reduction of 2.06 kW, or 56% between 7 and 8 am.

Supplemental MSHP Performance on Peak Summer Days
Figure 2. Time series data showing heating, ventilating, and air-conditioning energy use by air-conditioner compressor, air handler unit and strip heat, and supplemental mini-split for Site 60.

Supplemental MSHP Performance on Peak Winter Days

Supplemental MSHP Performance on Peak Winter Days
Figure 3. HVAC demand on FPL system peak winter day - 2014 vs. 2015.

In summary, the median annual HVAC energy reductions for the supplemental MSHP were impressive at 34%, with utility demand reductions of the supplemental mini-split also very large in both summer and winter, for the small sample of ten and six sites, respectively.

Reductions to long-term average interior RH were sometimes observed, albeit inconsistently.

Karen Sutherland is a research analyst at the Florida Solar Energy Center (FSEC). Danny Parker is a principal research scientist at the FSEC. Eric Martin is a program director at the FSEC. All three of them work in the Buildings Research Division.

Learn More

Barkaszi, S. Jr., and D. Parker. 1995. Florida Exterior Wall Insulation Field Test: Final Report. Florida Solar Energy Center (FSEC) FSEC-CR-868-95, Oak Ridge National Laboratory, Oak Ridge, TN (US).

Baylon, D., Larson, B., Storm, P., & Geraghty, K. 2012. “Ductless Heat Pump Impact & Process Evaluation: Field Metering Report” Ecotope, (Vol. #E12-237). Seattle, WA.

Brand, L. 1987. “Critical Needs Weatherization Research Project Final Report” Contract No. 5086-245-1352. Governor’s Energy Council, Harrisburg, PA,) Gas Research Institute.

Faesy, R., J. Grevatt, B. McCowan, and K. Champagne. 2014. “Ductless Heat Pump Meta Study” Northeast Energy Efficiency Partnership, Lexington, MA: NEEP.

Lubliner, M., L. Howard, D. Hales, R. Kunkle, A. Gordon, and M. Spencer. 2016. “Performance and Costs of Ductless Heat Pumps in Marine-Climate High-Performance Homes – Habitat for Humanity The Woods” National Renewable Energy Laboratory (NREL), Golden, CO.

Marshall, B., E. Swan. 2014. “Ductless Mini Splits Are Taking Off” Heating, Cooling & HVAC (post),, May 9, 2014.

Parker, D., K. Sutherland, D. Chasar, J. Montemurno, and J. Kono. 2014. “Measured Results of Phased Shallow and Deep Retrofits in Existing Homes.” Proceedings of the ACEEE 2014 Summer Study on Energy Efficiency in Buildings, 1:261–276. Washington, DC: ACEEE.

Parker, D., K. Sutherland, D. Chasar, J. Montemurno, and J. Kono. 2016. “Phased Retrofits in Existing Homes in Florida Phase I: Shallow and Deep Retrofits (February 2016)” National Renewable Energy Laboratory (NREL), Golden, CO.

Roth, K., N. Sehgal, and C. Akers. 2013. “Ductless mini-Split Heat Pump Comfort Evaluation” National Renewable Energy Laboratory (NREL), Golden, CO.

Sutherland, K., D. Parker, E. Martin, D. Chasar, and B. Amos. 2016. “Phased Retrofits in Existing Homes in Florida Phase II: Shallow Plus Retrofits (February 2016)” National Renewable Energy Laboratory (NREL), Golden, CO.

Welch, C. and Rogers, B, 2010.“Estimating the Remaining Useful Life in Appliances,” Proceedings of the 2010 Summer Study on Energy Efficiency in Buildings, 2:316–327. Washington, DC: ACEEE.

  • 1
  • NEXT
  • LAST
Click here to view this article on a single page.
© Home Energy Magazine 2020, all rights reserved. For permission to reprint, please send an e-mail to
Home Energy Pros
Discuss this article with other home performance professionals at Home Energy Pros!.

Add a new article comment!

Enter your comments in the box below:

(Please note that all comments are subject to review prior to posting.)


While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.

Related Articles
SPONSORED CONTENT Insulated, Air-Sealed Drapes Learn more! Watch Video